Comparison of NARX Neural Network and Classical Modelling Approaches

Abstract:

Classical optimization tools are effective when precise mechanistic models exist to support their design and implementation. However, most of the real-world processes are complex due to either nonlinearities or uncertainties (or both) and environmental variations, thus making realizing accurate mathematical models for such processes quite difficult or often impossible. Black box approach tends to present a better alternative in such situations. This paper presents a comparison of nonlinear autoregressive with eXogenous (NARX) neural network and traditional modelling techniques [autoregressive with exogenous input (ARX) and autoregressive moving average with exogenous input (ARMAX)]. The models were validated using experimental data from full-scale plants. Simulation results revealed that the performance of the NARX neural network is better compared to the ARMAX and ARX. The NARX neural network may serve as a valuable forecasting tool for the plants.

• Periodical: Applied Mechanics and Materials (Volume 554)
• Pages: 360-365
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.554.360
• Online since: June 2014
• Authors: Muhammad Gaya Sani, Norhaliza Abdul Wahab*, Yahya M. Sam, Sharatul Izah Samsudin, Irma Wani Jamaludin
• Keywords: Black Box, Estimation, Model, Neural Network (NN)

References

[1] O. Nelles, Nonlinear system identification: from classical approaches to neural networks and fuzzy models, Springer, Germany, (2000).

[2] M.S. Gaya, N.A. Wahab, Y.M. Sam, S. I Samsuddin, Feed-Forward Neural Network Approximation Applied to Activated Sludge System, Commun. Comput. Inf. Sci., 402(2013) 587–598.

DOI: 10.1007/978-3-642-45037-2_63

[3] C. Lewis, Industrial and business forecasting methods: A practical guide to exponential smoothing and curve fitting, Butterworth-Scientific, California, (1982).

[4] W. McCulloch, W. Pitts, A logical calculus of the ideas immanent in nervous activity, Bull. Math. Biophys., 5(1943) 115–133.

DOI: 10.1007/bf02478259

[5] A. Jain, J. Mao, and K. Mohiuddin, Artificial neural networks: A tutorial, IEEE-Computer Mag., 29 (1996) 31–44.

DOI: 10.1109/2.485891

[6] F. Karray, C. De Silva, Soft computing and intelligent systems design: theory, tools, and applications, Pearson Education Limited, England, (2004).

[7] T. Lin, C. Giles, A delay damage model selection algorithm for NARX neural networks, IEEE Transac. on Signal Process., 45(1997) 2719–2730.

DOI: 10.1109/78.650098

[8] Y.M. Chiang, L.C. Chang, F.J. Chang, Comparison of static-feedforward and dynamic-feedback neural networks for rainfall–runoff modeling, J. Hydrol., 290 (2004) 3–4.

DOI: 10.1016/j.jhydrol.2003.12.033

[9] A. Minns, M. Hall, Artificial neural networks as rainfall-runoff models, Hydrol. Sci. J., 41(1996) 399–418.

[10] R. Hecht-Nielsen, Kolmogorov's Mapping Neural Network Existence Theorem, in Proceedings of the First IEEE International Joint Conference on Neural Networks, New York, 1987, p.11–14.

[11] L. Rogers and F. Dowla, Optimization of groundwater remediation using artificial neural networks with parallel solute transport modeling, Water Resour. Res., 30(1994) 457–481.

DOI: 10.1029/93wr01494

[12] M.S. Gaya, N.A. Wahab N. A, Y.M. Sam , N. A Anuar , S. I Samsudin, A simplified model structure for an activated sludge system, Int. J. smart Sens. Intell. Syst., 6 (2013) 1167–1179. (2013).

DOI: 10.21307/ijssis-2017-585